Highlighting the crustal structure of southern Tuscany by the … › bgta › pdf ›...

425

Highlighting the crustal structure of southern Tuscany by thereprocessing of the CROP03 NVR profile

F. ACCAINO1, R. NICOLICH2 and U. TINIVELLA1,2

1 Istituto Nazionale di Oceanografia e di Geofisica Sperimentale – OGS, Trieste, Italy2 Dipartimento Ingegneria Civile, Università degli Studi di Trieste, Italy

(Received September 30, 2005; accepted April 5, 2006)

ABSTRACT. The formation of the Tyrrhenian basin and the formation and evolution of theApennines are believed to be the consequence of the settlement of two mantle domes,one in the southern Tyrrhenian Sea and one in the Tuscan-Latium area. These domesare considered the result of the interaction between the pre-existing lithosphericmantle (and crust) and a thermal input in presence of fluids derived from deep mantlesources. A consequence is the peculiar structure of the crust which interacts with theuppermost mantle, in a region where extensional regime, crustal thinning and magmaintrusions have been identified utilizing geophysical data, and in particular thereprocessing of reflection seismic lines crossing an area where important geothermalfields have been exploited. The new models were derived from heat-flow and gravityanomalies, deep seismic soundings, S-waves velocity models constructed by shallowand deep tomographic inversion and subsequent modelling of flow and stressdistribution in the lithosphere, deep reflection lines (CROP profiles). The reprocessedsection of the deep seismic profile CROP-03, in the sector from the Tyrrhenian shoreto the Siena graben, puts in evidence important features, that are compared directly tothe land profiles CROP-18A and -18B and to LISA lines acquired offshore, in theTuscan archipelago. The high resolution of reflection seismic images, and the AVOanalysis, highlight the lithology and physical properties of the crustal rocks ofsouthern Tuscany. The intrusion of mantle-derived magmas and lithologydifferentiation inside the lower crust, the vertical channels corresponding to the ascentpaths of magma from crust-mantle transition and fluids dominating the upper crust ofsouthern Tuscany represent the source and motor of the geothermal resources.

1. Introduction

Southern Tuscany is a key area for the exploitation of geothermal energy. The reservoirs werediscovered and studied from near-surface evidence and the geological contest and, more recently,utilizing 2D and 3D reflection seismic profiling, research furthered by ENEL, the main companyfor electric power supply in Italy.

High amplitude reflecting seismic horizon, generally followed by a remarkable and thick highenergy interval, was revealed within the metamorphic-crystalline units of the upper crust, at adepth of between 3 to 6 kilometres, when acquiring data over the main geothermal fields ofLarderello and Mount Amiata. It was indicated as the K-horizon by Batini et al. (1978) and itsunexpected reflectivity was mostly related to trapped fluids (occurrence of “bright spots”–like

Bollettino di Geofisica Teorica ed Applicata Vol. 47, n. 3, pp. 425-445; September 2006

© 2006 – OGS

426

Boll. Geof. Teor. Appl., 47, 425-445 Accaino et al.

images). However, any deep well sampled that horizon, because of the presence of severetemperatures and fluid pressure conditions (Batini et al., 1983). Therefore, the petrophysical andlithological properties of the K-horizon and the origin of high amplitude reflections within themetamorphic-crystalline basement cannot be proved directly.

To understand the crustal structures and their relationships to the geothermal fields threeseismic lines (CROP-18/A, -18/B and CROP-03, Fig. 1) were acquired on land in the frame ofthe CROsta Profonda (CROP) project [Scrocca et al., (2003) and references therein]. OtherCROP lines have been presented by Finetti (2005). Together with wide-angle reflection andrefraction deep seismic soundings (DSS) data (Giese et al., 1981), the CROP seismic lines giveinformation about the crustal structure of the geothermal province.

Several models have been recommended to describe the crustal process in the northernApennines and southern Tuscany. For example, Doglioni et al. [(2005), and references therein]suggest a subduction model, which can be passive. Locardi and Nicolich (2005) put forward that

Fig. 1 - Location map of CROP seismic lines acquired in Tuscany, from Elba Island to the Northern Apennines. Theanalysed sector of the profile CROP-03 is indicated as thick line. Traces of profiles crossing the Tuscan archipelagoare reported M-12, from Scrocca et al., (2003); Lisa-08, from Mauffret et al., 1999). The position of DSS profiles[N–S; C–B-A; from Giese et al., (1981)] is presented with dotted lines. Dots indicate also the position of the sketchmodel presented in Fig. 3.

427

Highlighting the crustal structure of S Tuscany Boll. Geof. Teor. Appl., 47, 425-445

after Middle Miocene the Tyrrhenian–Apennine region should be considered a closed kinematicsystem. Geodynamic forces were local and large lateral motions, induced by plate dynamics,were absent. They wrote that during the last 10-14 My these forces transformed a segment of theAlpine-Adriatic collision belt into a deep basin and into a new mountain chain. From the analysisof geophysical and geological data Locardi and Nicolich (2005) infer that mantle upwelling abovea deep-seated thermal plume caused this tectonic “revolution”. A support to this geologicaloccurrence is given by the proposed presence of a 5 to 20km thick layer of dense magma (a softmantle wedge, or mantle cushion) that underplates the western part of the Apennine range. Thislayer has contributed to the accretion of a new continental crust, an example of physical andchemical growing. The magmatic composition and the volcano-tectonic evolution indicate a rifttectonic environment and the transformation of the lithosphere mantle and crust by a thermalanomaly and by a fluid supply from deep mantle sources. A mantle plume model has beenrecently proposed also by Lavecchia et al. (2004) and Bell et al. (2004, 2005). Subduction (slabrollback or passive subduction) was widely discussed by Carminati et al. [(2005), with referencestherein].

Mapping Moho’s isobaths from DSS data, two distinct mantle domes, in the Tuscan – Latiumarea and in the South Tyrrhenian Sea, were defined (Locardi and Nicolich, 1988). The domesevolved independently and had different rates and direction of migration. After a first eastwarddrift, the northern one migrated to the northeast gaining a rate of up to 2.3 cm/y, whereas thesouthern dome moved to the southeast at a much higher rate of 6-8 cm/y (Locardi 1986, 1988).The different migration rates and directions can account for the different tectonic evolutionbetween the northern and southern Apennine range.

Fig. 2 shows a scheme for the mechanism of accumulation of fluids in a partly molten massat the base of the crust, considered across the southern dome, about 100 km south with respectto our investigated area. The fluidized sector acts both as lubricant and as driving force for thetectonic movements. The model is consistent with the convective motions denoted by: 1) theopening of the Tyrrhenian basin; 2) the migration of the oceanized zone; 3) the volcanism and 4)the tectonic transport (Locardi and Nicolich, 1988, 2005). In Tuscany, the above mentionedlowest migration rates of the northern dome prevented large lateral displacements of the crust-mantle system, but the new geophysical data discussed in this paper can detail a model also forthat region.

In Tuscany the volcanologic data show an eastward drift of the eruptive axes, since Tortonian,followed by their counterclockwise rotation, with the emplacement in its final position byPleistocene. Peccerillo (2002, 2003) proposes an ages interval spanning from 12-14 My, in thewestern margin of the Tuscan Archipelago, to about 0.2 Ma to the east. The almostcontemporaneous magmatic activity along single eruptive axes, hundreds of kilometres long, isanother feature of the northern dome. In fact, in the same area (Southern Tuscany, Latium,Campania), the high potash “Mediterranean” volcanism erupted along an inland arc of about a450 km length (Locardi, 1988) when the Apennine structure was already emplaced and its strongtectonic diversification along the range had already occurred. The eruption began almostcontemporaneously 0.9–0.6 My ago and can be considered presently active and along the wholearc.

Peccerillo (2003) and Peccerillo and Lustrino (2005) suggest a completely different model to

428


explain the origin of Tuscan magmatism, also taking into account the interaction between crustand mantle and the consequent geodynamic contest.

Locardi and Nicolich (2005) infer a link between crustal anatectic and high potash mantlemagmatism, which should correspond only to the area underplated by the heavy mantle material.They consider that the crust was molten above the migrating mantle dome where CO2-dominatedfluids exhaling from the mantle where it is accumulated.

Aoudia et al. (2004, 2005) inspected the crust and uppermost mantle evaluating thecontinental deformation and modelling the contemporary flow and stress distribution in thelithosphere. They investigated the contribution of internal buoyancy forces using the S-wavevelocity model by Chimera et al. (2003), retrieved with the CROP-03 profile data and with a high

Fig. 2 - Evolutionary scheme proposed for the asthenosphere diapir in the southern Tyrrhenian [a sketch transect fromSardinia Island to the Adriatic coasts of Apulia, from Locardi and Nicolich, (2005)], about 100 km south with respectto our investigated area. M1= Adria Moho, M2=Tyrrhenian exposed mantle, M3= top of mantle derived underplatedmaterial. Thin arrows= crustal structure delamination mechanism. Thick arrows: convective movements inside themantle. 1= crust reduced by delamination and successively melted; 2= zone of maximum fluids accumulation; 3=oceanization; 4= mantle-derived underplated material; 5= seismic arrays corresponding to Benioff’s zone of thesouthern Tyrrhenian reaching more than one hundred km at depth ; 6= isostatic uprising.

429


resolution gravimetric survey utilized to estimate density values (Marson et al., 1998) as a prioryconstraint of shallow and deep tomographic inversion of surface waves. A sharp well-developedlow-velocity zone in the uppermost mantle (mantle wedge, cushion), between 30 and 50 km depthand more than 130 km wide, from the Tyrrhenian dying out beneath the Apennines, separatescrust and lid (sketch model in Fig. 3).

The puzzling hypocenter location of the sub-crustal seismicity, which does not define aWadati-Benioff seismic zone (Selvaggi and Amato, 1992; Barchi et al., 2006), is explained byAoudia et al., (2005), by the particular geometry and buoyancy of the delaminating lithosphere.In the crust, their model predicts compression regime at the eastern part of the profile and tensionin Tuscany over the mantle wedge with severe earthquakes in the region where the lithosphere ishighly deformed.

Geophysical data acquired across the Tuscan geothermal province support this model.Important information is represented by refraction and wide-angle reflection deep seismicsoundings [DSS; Giese et al., (1981)], which defined the top of the lower crust at a 12 to 16 kmdepth, marked by an interface with a refraction velocity of 6.8 – 7.3 km/s. The Mohodiscontinuity is at 22-25 km depth (Nicolich, 1989; Locardi and Nicolich, 2005), characterizedby an anomalous low-interval velocity of 7.5-7.8 km/s for the culmination of the above mantlewedge beneath the geothermal province. Simplified crustal models along two DSS profiles (N-S directed NW-SE and C-B-A, WSW-ENE) are reported in Fig. 4. The lower part of the crust is

Fig. 3 - Sketch of the lithosphere-asthenosphere system, with a soft mantle wedge beneath Tuscany, approximatelyfollowing the CROP-03 transect [after Aoudia et al., (2004); position in Fig. 1].

430


marked and has been defined as anomalous with the presence of high and low velocity layers.The rapid velocity changes can favor the observed reflectivity at DSS’s wavelengths. The velocitymodel was refined by Ponziani et al. (1995) and by De Franco et al. (1998) mantaining thedifferentiation of the crustal layers down to the Moho.

Seismic velocities, that are intermediate between the velocities of the mantle and those of thelower crust (7.5-7.8 km/s) are generally interpreted as underplating of heavy mafic meltsextracted by a hot mantle plume (Bonatti and Seyer, 1987). The area where those values havebeen found is delimited by the highest heat-flow values (Della Vedova et al., 1991, 2001), thatcorrespond to the extension of the mantle wedge and are in agreement with an upward flow field,suggesting that the mantle wedge is partially molten.

Gravity data, when the Bouguer anomalies are computed with a reduction density of 2.670kg/m3, are evidence of an excess of mass in the lower crust of the geothermal province, which isisostatically undercompensated (Velicogna et al., 1996; Marson et al., 1998). In addition, thegeoid height shows a steep gradient moving from western Tuscany towards the Umbria-Marcheregions (Barzaghi et al., 2002). On the contrary, local scale negative anomalies, when utilizing adensity of 2.400 kg/m3, indicate a deficit of mass in the upper crust (Nicolich and Marson, 1994;Tinivella et al., 2005).

Pluton emplacements, studied by Acocella and Rossetti (2002), require roof uplift to provide

Fig. 4 - Interpretation scheme of DSS data (from Giese et al., 1981) along the C-B-A and N-S profile (position in Fig.1).

431


the pathway to the ascent of granite magmas in the brittle-extending crust to give the necessaryspace to the ascent and emplacement of the volcanic staff. A broad uplift of more than 600 m forPliocene deposits characterize the geothermal fields. A geological and tectonic description of thesouthern Tuscany can be found in Decandia et al. (1988), Liotta et al. (1998), Brogi et al. (2003,2005), with references therein.

In this paper, we present the seismic images of the deep crustal structures corresponding tothe geothermal fields of Tuscany and the results of non-conventional analyses (such as, trueamplitide processing, pre-stack migration and amplitude versus offset). We stress that theresolution and the penetration of the crustal data cannot provide a detailed description of the verycomplex near-surface geology and of the mantle structures.

2. CROP-03 reprocessed seismic data

The recent deep seismic reflection data, recorded in the studied area [CROP-03 and CROP-18A, B profiles: Pialli et al., (1998); Accaino et al., (2005b)], are of great interest. The results ofreprocessing and analysis of the available data of the western sector of the seismic line CROP-03(indicated with a thick line in the map of Fig. 1) are presented through a multidisciplinaryapproach here. The main objective of the study was determined by the need to obtain thestructure of the crust completed by information on the composition and fluids content. Improvingthe signal/noise ratio and preserving the reflection amplitudes, the distinct seismic units aredifferentiated offering a direct image of the variable lithological and petrophysical properties

Fig. 5 - Stack section after true-amplitude processing of the western portion of the CROP-03 profile (see Fig. 1). Datumplane: + 200 m above sea level. The ratio between P- and S-waves reflectivity is over-imposed to the section at selectedCDP (see the text for details). An arrow indicates the crossing point of the line CROP-18B. The horizontal solid lineshows the overlapping with the receivers of the wide-angle acquisition with shots fired in correspondence of the blackcircle.

432


characterizing the crust of the geothermal province. The applied processing flow chart is consistent with the elaboration performed on the other

CROP lines in the area [CROP-18A and -18B; details in Accaino et al., (2005b)] allowing thecomparison of different seismic sections. Shot/receiver geometry applied to the raw data definedthe common mid points (CMPs) producing correctly stacked signals. Surface-consistentamplitude balancing completed the recovery of the amplitude losses (spherical divergence andabsorption phenomena or variable ground coupling of the geophones and diverse energy of eachshot). Source and receiver static corrections were evaluated, after picking the first arrivals, bymeans of the commercial software (GEOTOMO©) that utilises tomographic approaches. Surfaceconsistent deconvolution has been applied selecting a prediction lag of 24 ms (that correspondsto the second zero crossing) and an operator length of 140 ms. A preliminary velocity analysis,after a trimmed dip filter (Holcombe and Wojslaw, 1992), with the data corrected to a floatingdatum, has been evaluated. With this procedure we have removed the ground roll with a

Fig. 6 - Stacked sections of the lines CROP-18B (Accaino et al., 2005b) and CROP-03 plotted in perspective. Thecontinuity between the two sections in correspondence with the main reflection zone starting at about 4 s TWT isevident. The K-horizon reflecting interval is imaged only beneath the Mount Amiata geothermal field.

433


satisfactory result. The sequence preserved the amplitude spectrum significantly. To improve thesignal-noise ratio, F-X deconvolution was applied in the shot domain after normal moveout(NMO) and field static corrections. Removing NMO and statics, a new stack velocity analysiswas carried out. A refined NMO correction and residual statics, using a surface consistentprocedure, were then computed and applied. Finally, a time variant filter and AGC, with a timewindow of 3 s, completed the stack section shown in Fig. 5.

The section is also plotted in perspective with the seismic line CROP-18B [Fig. 6; Accaino etal., (2005b)] to put in evidence the correspondence of the main reflecting intervals recognisableon both lines. The high reflectivity, below 5 s TWT, is present in both sections, while the K-horizon, characterizing the neighbouring Mount Amiata geothermal field cannot be followed onthe line CROP-03. The re-processing confirms a quite flat reflector around 4 s, which can becorrelated with the regional interface already considered on CROP-18A and B, below the K-horizon, and associated to a lithologic change (Tinivella et al., 2005).

The main reflectivity along the CROP-03 section can be identified directly in theinstantaneous amplitude section (Fig. 7), presented after the seismic attributes analysis. Asdistinctly evidenced, the deeper part (from 5 to 8 s TWT) provides reflections with much higherenergy compared with the shallow part (above 4 s). The Moho is marked by the termination ofthe high-amplitude continuous, fairly thick reflecting intervals at 7.5-8 s TWT. Somewhere a lackof reflected energy occurs, like the transparent vertical corridors crossing the lower section. Theshallow non-reflecting zones can be explained by the presence of energy scattering by multiplemagma intrusions.

Fig. 7 - Instantaneous amplitude obtained from stacked section with the highlighted image of the lower crust and withblue transparent zones related to magma intrusions through the lower crust and within the upper crustal layers.

434


3. Evaluation of the seismic velocities

The velocity analysis was performed by using iteratively the pre-stack migration, employingthe commercial software GeoDepth©. At the first inversion step, we considered a velocity modelwith a constant velocity, equal to 2500 m/s. Thus the pre-stack time-migration is computed andthen the Common Image Gathers (CIGs) were analysed, determining the residual move-outcorrections from the semblance analysis. The model is updated by using the following rule: theflatness of an event on CIG yields a qualitative estimation of the error in the determination of thevelocity model (Yilmaz, 2001). The residual moveout analysis of the CIGs and the update of thevelocity model were iteratively repeated until the residual moveout was near zero. We used alayer-stripping approach, updating the velocity following the time sequence. In this way, velocityversus time is better constrained. The main advantage of using this iterative procedure is that thepicking in the pre-stack domain is not necessary, particularly difficult in the case of poor land-recorded data.

The deeper velocity information (below 5 s TWT) cannot be obtained, because the maximumavailable offsets do not allow a detailed move-out analysis. So, we introduced a velocity gradientdetermined (i) from the velocities identified by refraction data (Giese et al., 1981), and (ii)upgrading the seismic imaging of the pre-stack time-migration. The criterion followed was toobtain the same depth for the base of the crust in the reflection section image as defined by DSS

Fig. 8 - Root-mean-square velocity distribution used for the pre-stack time-migration. The arrow indicates the positionof the line CROP-18B. A comparison between the RMS-velocity from the profile CROP-18B (dotted line) and CROP-03 (solid line) is presented in correspondence to their crossing.

435


(Fig. 3), also after tests performed on an acceptable velocity-dependent migration of the data.Note that Fig. 3 reports interval velocities in each layer, whereas Fig. 8 reports root mean square(RMS) velocity functions . The two velocities are related by Dix’s equation:

where Vint is the interval velocity within the layer boundaries n and n-1; Vn and Vn-1 are the RMSvelocities down to the layer boundaries n and n-1, respectively, and tn and tn-1 are the two-waystimes to these boundaries.

The RMS velocity model (Fig. 8), obtained by the above iterative procedure, was smoothedand a pre-stack time-migration computed (Fig. 9a), improving the correct positioning of theseismic horizons and of dipping anomalous events.

In the lower window (Fig. 9b) the line-drawing is shown. The details of the shallow part (aboutthe first 3 km depth) can be obtained by using wells and available geological information, which

V t t = V t V tn n n n n nint2

12

12

1−( ) −( )− − − ,

Fig. 9 - a) pre-stack time-migration; b) line-drawing. The crosses indicate the possible magmatic intrusions associatedwith low reflecting energy. The line-drawing of the upper crust was not controlled with the outcroppings and is onlyindicative of the brittle nature of that part. The main reflectors in the lower crust are indicated with thick segments. TheMoho boundary corresponds to the lower termination of the lower crust lamellae. The arrow indicates the location ofthe line CROP-18B.

a

b

436


are exhaustively described in Decandia et al. (1998). The termination of layered reflectivities within the lower crust is considered to coincide with

the Moho discontinuity at about a 23 km depth. The reprocessed data of the CROP-03 section do not offer the interpreter any seismic image

that can allow the definition of the very intriguing features presented in the paper of Finetti et al.(2001). On the contrary, our data match those recorded in the Tuscan Archipelago by Mauffret etal., (1999) very well (LISA project). In Fig. 10, the line drawing of the LISA-08 line, whichcontinues inside the Tuscan Archipelago domain the lower crust structures made known on land.

In Fig. 10, the line drawing of the LISA-08 line, continue inside the Tuscan Archipelagodomain the lower crust structures pointed out on land.

We already mentioned that the seismic section reveals the presence of sub-vertical conduits orfractures that can be related to the ascent of plutons from the lower to the upper crust, identifiedalso on CROP-18B and -18A (Accaino et al., 2005b; Tinivella et al., 2005). To better evidencethis aspect, we present a pre-stack depth-migrated section by using the final velocity model in thesector between the CDPs 1100 and 2100. Fig. 11a indicates the section where the opposite dip ofthe reflectors with respect to a vertical channel is marked Fig. 11b shows the similar feature, asrevealed on the CROP-18B (from Tinivella et al., 2005).

4. AVO analysis

Both the stacked section and the pre-stack time-migration sections reveal the presence of aquite flat reflector at a depth (~ 9 km – 4 s TWT) already recognised in the other seismic linesacquired in Southern Tuscany (Accaino et al., 2005a, 2005b; Tinivella et al., 2005). To identify if

Fig. 10 - Line-drawing of the profile LISA-08 [from Mauffret et al., (1999); position in Fig. 1]. Note the dippingreflection in the eastern part of the section. The lower crust interval matches the onshore data of CROP-03. The Mohois located at about 7.5-8 s TWT.

437


the reflections are related to the lithology or to pore fluids changes, we completed an amplitudeversus offset (AVO) analysis by using the GeoDepth© (Paradigm Geophysical) commercialsoftware. We followed the Aki and Richards’s (1980) approach for the linearization of theZoeppritz equations. The main inputs are the pre-stack true amplitude data, the velocity model,and the acquisition geometry, while the outputs are two sections that reported the P- and S-wavereflectivities, linked to the main P- and S-wave velocity contrasts. For more details about theprocedure see Tinivella et al., (2005).

The P- and S-wave reflectivity sections are shown in Fig. 12 and reveal (i) the presence of a

Fig. 11 - a) Detail of the pre-stack depth-migration in the central part of the line CROP-03 as recognizable in this figureand in Figs. 7, 9 or 12 looking at CDP numbers. b) image of a vertical channel obtained on the line CROP-18B(Accaino et al., 2005a).

a

b

438


low S-energy above 4 s TWT [probably related to fluids in the pore space in agreement withTinivella et al., (2005)], and (ii) the relatively high-energy reflectors below 5 s TWT in bothsections. Thus, the AVO results confirm the absence of important lithologic changes and supportthe presence of fluids above 4 s TWT.

To help the interpretation of the AVO results, we evaluated the ratio between the P-wave and S-wave velocity reflectivity sections, providing information about the Poisson ratio contrast. The resultsare reported at selected CDPs in the stacked section (see Fig. 5) and indicate the presence of thereflector at about 4 s TWT with high P/S reflectivity ratio most likely caused by a lithologic change.

Fig. 12 - P-waves reflectivity (upper section) and S-waves reflectivity (lower section) after AVO inversion. The arrowindicates the intersection with the line CROP-18B. Note the low reflection zone in the upper part of the two sectionsand the energy present in both sections at about 4 s TWT.

439


5. Wide-angle shotsDuring the acquisition of the seismic line CROP-03 several high energy shots were used to

record expanding spread data (Gualtieri et al., 1998). In this paper we consider three shots locatedat CDP 2885 position (Fig. 5) and recorded until 40 km of distance. The details of the acquisitiongeometry are reported in Table 1. Because different source energy was used during the threeacquisitions, we considered the overlapping traces to rescale and equalize the amplitudes (Fig.13). The reflector at about 4 s TWT is recognizable at near offsets.

Picking this reflector we extracted the amplitudes, averaging signal within a window wide0.05 s. The amplitudes were rescaled applying a geometrical spreading correction considering the

Dynamite (kg) First offset (m) CDP at first offset Last offset (m) CDP at

last offset

80 5412 2706 16,837 2326

150 16,537 2335 27,994 1954

200 27,694 1963 39,051 1586

Table 1 - Acquisition geometry of the three wide-angle shots.

Fig. 13 - Wide-angle records after amplitude compensation (see text for details). The position of shot and receivers areindicated in Fig. 5. The dotted line points to the picked reflector. The insert shows the energy versus incidence angleextracted from the data (stars) and the best theoretical fit (continuous line).

440


effects of the offsets (Ursin, 1990) and utilizing the velocity model obtained for the line CROP-03. The offset was converted into incidence angles by using the velocity information. Finally, theamplitudes versus incidence angles were rescaled with respect to the maximum amplitude andfiltered. Note that the extracted amplitudes are not precisely related to the reflection coefficientsat the selected reflector, because the reflection points at different offsets are broadly distributedalong the reflecting interface. Observing the geometry (quite flat) and the reflectivity of theinterface (quite constant strength amplitude, as evidenced in Fig. 7) along the seismic line, we cansuppose that the assumption of the coincidence uncertainty of the reflection point at all offsetshas the same order of magnitude of other errors included in the analysis, such as the velocitymodel. These amplitudes are related to the contrasts of the physical parameters that caused thereflection, i.e. the P- and S-waves velocities and density, by using a theoretical model forviscoelastic isotropic media [single-phase model: Carcione, (1997)]. The algorithm fordetermining these parameters is the least-square method; we estimated the physical parametercontrasts minimising the function that represents the square difference between the amplitudesobtained from seismic data and those obtained from the theory. The results (see the insert in Fig.13) indicate that P-wave reflectivity is controlled by density variation, that is by the presence ofchanges of lithology.

6. Discussion

The comparison of the velocity functions of CROP-03 and CROP-18B (Fig. 8) in proximityof their crossing point indicates the excellent match between the two profiles above 5 s TWT,confirming the goodness of the two approaches (tomographic inversion, used in CROP-18, andCIG analysis). On the contrary, below 5 s the two velocity profiles present strong differences.Because the two seismic lines are dip (CROP 03) and strike (CROP 18B) lines with respect to themain regional structures (Apenninic and anti-Apenninic), the velocity difference can be relatedto anisotropy [see also Margheriti et al., (1996)]. The presence of anisotropy in the lower crusthas been evidenced by many authors [see for example the recent papers of Godfrey et al., (2002),Shapiro et al., (2004), Wilson et al., (2004), Lueschen, (2005)]. Weiss et al. (1999) point out thatthe seismic anisotropy is particularly strong in layered structures, such as the seismic lamellaethat are typical of the lower crust. They reached the conclusion that the deformation that occursby slip on discrete faults in the upper 15 km of the crust is distributed as strain over aprogressively broader region below, a typical situation for Tuscany. Millahn et al. (2005) havefound in the Tauern Window (Transalp profile crossing the Eastern Alps) a 10% velocityanisotropy attributed to E-W elongation of the rock texture caused by paleo-stresses, that is N-Scompression and E-W extrusion and stretching.

In our case, the percentage anisotropy (A) is between 10 and 20%, calculated by using thefollowing relationship:

where V03 is the maximum velocity in the WSW-ENE direction and V18B (NW-SE direction) isthe minimum velocity. Seismological studies already revealed the presence of anisotropy in

A V V V= ( - )/03 18B 03

441


southern Tuscany (Margheriti et al., 1996) with an approximately E-W orientation of the high-velocity direction attributed to the stretching of the upper mantle below the Tyrrehenian Sea.

The seismic sections and the AVO analyses on both wide-angle and near-vertical reflectionseismic confirm the presence of a regional reflector at about 4 s (i.e., ~ 9 km depth) that can bepositioned at the base of a more transparent interval. The AVO inversion confirmed also theoccurrence of non-reflecting sectors that can be justified accepting the occurrence of magmaascent along vertical conduits within the lower crust, and of energy scattering by intrusions in theupper crust. The presence of overpressured trapped fluids can explain the reflectivity in the uppercrust (K-horizon) revealed on the neighbouring geothermal fields (Tinivella et al., 2005).

Dini et al. (2005) presented the assumption that crustal sources for magmas (granites) arelocated at 14-23 km depth (5 to 8 s TWT). They estimated an efficient, fast magma extractionfrom that source, while mantle derived magmas are supposed to be present at the base and in thelower crust (again the interval between 14 and 23 km).

Heat flow sources are confined to at depths of 3 km or less, contributing to the very highvalues in correspondence of the geothermal fields and at short wavelengths: local advection ofhot fluids. To take account of regional conductive heat transfer with larger wavelengths, thecontribution of deep sources (at 8 km or larger) is obligatory (Bellani and Della Vedova, 2003).

The intrusion of mantle-derived magmas is commonly referred to as magmatic underplatingand has been studied by Sinigoi et al. (2003, and references therein) along the considerableoutcroppings of lower crust bodies in the Ivrea-Verbano zone (Southern Alps in NW Italy). Theheat released at the base of the crust induced anatexis in the overlying crustal rocks withproduction of granitoid melts, quickly migrating towards higher crustal levels, leaving behindprogressively depleted restites, the residual product of the fusion processes, representing high-grade metamorphic and refractory rocks. A sharp distinction between denser materials and lessdense anatectic melts and metamorphosed volcano-sedimentary rocks is the end result.

Seismic images can support the model suggested by Serri et al. (1993), i.e., the partial meltingof the uppermost mantle and the delamination processes, even if they relate it to the subductionprocesses of the Adriatic continental lithosphere (Doglioni et al., 1999, 2005). Our analysis ofseismic data does not support the existence of active subduction zones activated by externalconvergence forces.

Upper mantle mafic intrusion in the lower crust layers is consistent with volcanologicalobservations as well as with reflection seismic images and with a layered lower crust marked byvelocity (and density) alternations in the DSS profiles. In the upper crust volcanic products,fluids and overpressures are accepted by our analysis. From the theoretical modelling, theporosity in the overpressured zones turn out to have a value near 5.5% (Tinivella et al., 2005),with a consequent decrease of the density that can partly justify the negative gravity Bougueranomalies observed in correspondence of the main geothermal fields.

7. Conclusions

The formation of the Tyrrhenian basin and the formation and evolution of the Apennines arethe effect of two asthenospheric mantle domes. These domes should be considered, to a largeextent, as a physical and chemical transformation of the pre-existing lithospheric mantle. The

442


magma formation was induced by heat and mantle metasomatic fluid supply from deep sources;it is produced at the top of the ascending mantle plume and is accreted at the base of the crust-underplating or continental crust accretion in a rift tectonic environment.

In the southern Tuscan delamination of the uppermost lithosphere beneath the base of the crustis observed by tomographic inversion of surface S-waves from seismological data, defining thepresence of a mantle wedge. Modelling shows that lithospheric deformations are driven bybuoyancy forces interrelated with this mantle wedge.

In conclusion, our integrated analyses of geophysical data are in agreement with the abovedescribed hypothesis and, in particular, we can recognise in the seismic images presented here therising of hot melts from below, with unstable equilibrium among rock physical properties,temperatures, fluid pressures and confining pressures at different levels of the crust, source andmotor of the geothermal energy resources.

Acknowledgements. The authors wish to thank Michela Giustiniani for the tomographic inversion withGEOTOMO and static corrections along the CROP-03 profile. Thanks are given to Massimiliano Barchifor comments and helpful reviews. An anonymous reviewer is acknowledged.

REFERENCESAccaino F., Tinivella U., Rossi G. and Nicolich R.; 2005a: Geofluid evidence from analysis of deep crustal seismic data

(Southern Tuscany, Italy). In: Liotta D. and Ranalli G. (eds), Structures in the Continental Crust and GeothermalResources, J. of Volc. and Geotherm. Res., Elsevier, The Netherland, 148, 46-59.

Accaino F., Tinivella U., Rossi G. and Nicolich R.; 2005b: Imaging of CROP-18 deep seismic crustal data. Boll. Soc.Geol. It., 3, 195-204.

Acocella V. and Rossetti F.; 2002: The role of extensional tectonics at different crustal levels on granite ascent andemplacement: an example from Tuscany (Italy). Tectonophysics, 154, 71-83.

Aki K. and Richards P.G.; 1980: Quantitative seismology. Theory and methods,1. W.H.Freeman and Co., SanFrancisco, 932 pp.

Aoudia A., Ismail-Zadeh A.T. and Panza G.F.; 2004: The simultaneous use of CROP and surface wave tomographyevidences lithospheric delamination and buoyancy-driven deformations in Central Italy. In: 32nd Intern.Geological Congress, Florence, Abstract 45-7. Cd rom.

Aoudia A., Ismail-Zadeh A.T. and Panza G.F.; 2005: Buoyancy-driven deformations and contemporary tectonic stressin the lithosphere beneath north-central Italy. J. of Geophys. Res., Submitted.

Barchi M.R., Pauselli C., Chiarabba C., Di Stefano R., Federico C. and Minelli G.; 2006: Crustal structure, tectonicevolution and seismogenesis in the Northern Apennines (Italy). Boll. Geof. Teor. Appl., 47, xx-xx.

Barzaghi R., Betti B., Borghi A., Sona G. and Tornatore V.; 2002: The Italian quasi-geoid ITALGEO99. Boll. diGeodesia e Scienze Affini, 1, 33-51.

Batini F., Burgassi P.D., Cameli G.M., Nicolich R. and Squarci P.; 1978: Contribution to the study of the deeplithospheric profiles: “deep” reflecting horizons in Larderello-Travale geothermal field. Mem. Soc. Geol. It., 19,477-484.

Batini F., Bertini G., Bottai A., Burgassi P.D., Cappetti G., Gianelli G. and Puxeddu M.; 1983: San Pompeo 2 deep well:a high temperature and high pressure geothermal system. In: European Geothermal Update, 3rd Intern. Seminar ECRes. Demonstration Projects, Ext. Abs., pp. 341-353.

Bell K., Castorina F., Lavecchia G., Rosatelli G. and Stoppa F.; 2004: Is there a mantle plume below Italy? EOS, 85,541-547.

Bell K., Lavecchia G. and Stoppa F.; 2005: Reasoning and beliefs about Italian geology. Boll. Soc. Geol. It., 5, 119-127.

443


Bellani S. and Della Vedova B.; 2003: Structures in the continental crust and geothermal resources. In: Brogi A.,Ghinassi M. and Liotta D. (eds), Geothermal Congress, Abs. vol., Univ. of Siena, Tipografia Senese, Siena (Italy),pp. 34-35.

Bonatti E. and Seyler M.; 1987: Crustal underplating and evolution in the Red Sea rift. J. Geophys. Res., 92, 12083-12121.

Brogi A., Liotta D., Lazzarotto A., Ranalli G. and Nicolich R.; 2003: L’orizzonte K nella crosta di Larderello (Campigeotermici della Toscana meridionale). Boll. Soc. Geol. It., 122, 103-116.

Brogi A., Lazzarotto A., Liotta D., Ranalli D. and CROP-18 Working Group; 2005: Crustal structures in the geothermalareas of southern Tuscany (Italy): insights from CROP-18 deep seismic reflections lines. In: Liotta D. and RanalliG. (eds), Structures in the Continental Crust and Geothermal Resources. J. of Volc. and Geother. Res., ElsevierSience, The Netherland, pp. 60-80.

Carcione J.M.; 1997: Reflection and transmission of qP-qS plane waves at a plane boundary between viscoelastictransversely isotropic media. Geophys. J. Int., 129, 669-880.

Carminati E., Negredo A.M., Valera J.L. and Doglioni C.; 2005: Suduction-related intermediate deep seismicity inItaly: insights from thermal and rheological modelling. Phys. of Earth and Plan. Int., 149, 65-79.

Chimera G., Aoudia A., Saraò A. and Panza G.F.; 2003: Active tectonics in central Italy: constraints from surface wavetomography and source moment tensor inversion. Phys. Earth Planet. Int., 22, 241-262.

Decandia F.A., Lazzarotto A., Liotta D., Cernobori L. and Nicolich R.; 1988: The CROP-03 traverse: insights on post-collisional evolution of Northern Apennines. Mem. Soc. Geol. It., 52, 427-439.

De Franco R., Ponziani F., Biella G., Boniolo G., Caielli G., Corsi A., Maistrello M. and Morrone A.; 1998: DSS-Warexperiment in support of the CROP-03 Project. Mem. Soc. Geol. It., 52, 67-90.

Della Vedova B., Marson I., Panza G.F. and Suhadolc P.; 1991: Upper mantle properties of the Tuscan-Tyrrhenian area:a key for understanding the recent tectonic evolution of the Italian region. Tectonophysics, 195, 311-318.

Della Vedova B., Bellani S., Pellis G. and Squarci P.; 2001: Deep temperature and surface heat flow distribution. In:Vai G.B. and Martini I.P. (eds), Anatomy of an Orogen – The Apennines and adjacent Mediterranean Basins, KluverAcad. Publishers, Dordrecht, Boston, London, pp. 65-76.

Doglioni C., Guenguen E., Harabaglia P. and Mongelli F.; 1999: On the origin of W-directed subduction zones andapplications to the Western Mediterranean. Geol. Soc. Special Publ., 156, 541-561.

Doglioni C., Cuffaro M. and Carminati E.; 2006: Some rearks on subduction zones. Boll. Geof. Teor. Appl., in press.

Dini A., Giannelli G., Puxeddu M. and Ruggieri G.; 2005: Origin and evolution of Pliocene-Pleistocene granites fromthe Larderello geothermal field (Tuscan Magmatic Province, Italy). Lithos, 81, 1-31.

Finetti I.R. (ed); 2005: CROP Project: deep seismic exploration of the Central Mediterranean and Italy, Elsevier B.V.,Trento, 798 pp.

Finetti I.R., Boccaletti M., Bonini M., Del Ben A., Geletti R., Pipan M. and Sani F.; 2001: Crustal section based onCROP seismic data across the North Tyrrhenian – Northern Apennines - Adriatic Sea. Tectonophysics 343, 135-163.

Giese P., Wigger P., Morelli C. and Nicolich R.; 1981: Seismische Studien zur Bestimmung der Krustenstruktur-anomalien der Toskana. Comm. Europ. Communities, EUR, 108 pp.

Godfrey N.J., Christensen N.I. and Okaya D.A.; 2002. The effect of crustal anisotropy on reflector depth and velocitydetermination from wide-angle seismic data: a synthetic example based on South Island, New Zealand.Tectonophysics, 355, 145-161.

Gualtieri L., De Franco R. and Mazzotti A.; 1998: A velocity model along the CROP-03 profile derived from expandingspread experiments. In: Pialli G., Barchi M. and Minelli G. (eds), Results of the CROP-03 deep seismic reflectionprofile, Mem. Soc. Geol. It., 52, pp. 139-152.

Holcombe, H. T., and Wojslaw, R. S.; 1992. Spatially weighted trim stacking: a technique for pre-stack noisesuppression. In: Proceedings Society of Exploration Geophysicists, Annual Meeting, pp. 1157-1160.

Lavecchia G., Boncio N., Creati N. and Brozzetti F.; 2004: Some aspects of the Italian geology not fitting with asubduction scenario. Virtual Explorer, 10, 1-14.

Liotta D., Cernobori L. and Nicolich R.; 1998: Restricted rifting and its consistence with compressional structures:results from CROP-3 traverse (Northern Apennines, Italy). Terra Nova, 10, 16-20.

Locardi E.; 1986: Tyrrhenian volcanic arcs: volcano-tectonics, petrogenesis and economic aspects. In: Wezel F.C. (ed),

444


The origin of the arcs, Elsevier Sc. Publ., Amsterdam, pp. 351-373.

Locardi E.; 1988: The origin of Apenninic arcs. Tectonophysics, 146, 105-123.

Locardi E. and Nicolich R.; 1988: Geodinamica del Tirreno e dell’Appennino Centro-Meridionale: la nuova cartadella Moho. Mem. Soc. Geol. It., 41, 121-140.

Locardi E. and Nicolich R.; 2005: Crust-mantle structures and Neogene-Quaternary magmatism in Italy. Boll. Geof.Teor. Appl., 46, 169-180.

Lueschen E.; 2005: Relatioship between recent heat flow and seismic properties: some notes from crustal research inGermany. J. of Volc. and Geother. Res., in press.

Marson I., Cernobori L., Nicolich R., Stoka M., Liotta D., Palmieri F. and Velicogna I.; 1998: CROP03 profile: ageophysical analysis of data and results. Mem. Soc. Geol. It., 52, 123-137.

Margheriti L., Nostro C., Cocco M., and Amato A.; 1996: Seismic anisotropy nemeath the Northen Apennines (Italy)and its tectonic implications. Geophys. Res. Lett., 23, 2721-2724.

Millahn K., Lueschen E., Gebrande H. and TRANSALP Working Group; 2005: TRANSALP - cross-lines recordingduring the seismic reflection transect in the Eastern Alps. In: Gebrande H., Castellarin A., Lueschen E., NeubauerF. and Nicolich R. (eds): TRANSALP- a Transect Through a Young Collisional Orogen. Tectonophysics, Elsevier,The Netherland, pp. 39-49.

Mauffret A., Contrucci I. and Brunet C.; 1999: Structural evolution of the Northern Tyrrhenian Sea from new seismicdata. Marine and Petroleoum Geology, 16, 381- 407.

Nicolich R., 1989. Crustal structures from seismic studies in the frame of the European Geotraverse (southernsegment) and CROP project. In: Boriani A., Bonafede M., Piccardo G.B. and Vai G.B. (eds), The Lithosphere inItaly, Accad. Naz. of Lincei, Roma, pp. 41-61.

Nicolich R. and Marson I.; 1994: Caratteri geofisici delle strutture crostali nella provincia geotermica Toscana. In:Lazzaretto A. and Liotta D., (eds), Studi preliminari all’acquisizione del profilo CROP-18, Larderello – M.teAmiata. Studi Geologici Camerti, Ediment, Città di Casello (Italy), pp. 163-168.

Peccerillo A.; 2002: Quaternary magmatism in central-southern Italy: a new classification scheme for volcanicprovinces and its geodynamic implications. Boll. Soc. Geol. It., 1, 113-127.

Peccerillo A.; 2003: Plio-Quaternary magmatism in Italy. Episodes, 26, 222-226.

Peccerillo A. and Lustrino M.; 2005: Compositional variations of Plio-Quaternary magmatism in the circum-Tyrrhenian area: deep versus shallow mantle processes. In: Foulger G.R., Natland J.H., Presnall D.C. and AndersonD.L. (eds), Plates, plumes and Paradigms. Geological Society of America, 388, pp. 421-434.

Pialli G., Barchi M. and Minelli (eds); 1998: Results of the CROP-03 deep sismic reflection profile. Mem. Soc. Geol.It., 657 pp.

Ponziani F., De Franco R., Minelli G., Biella G., Federico C. and Pialli G.; 1995: Crustal shortening and duplicationof the Moho in the Northern Apennines: a view from seismic refraction data. Tectonophysics, 252, 391-418.

Scrocca D., Doglioni C., Innocenti F., Manetti P., Mazzotti A., Bertelli L., Burbi L. and D’Offizi S. (eds); 2003: CROPATLAS – Seismic Reflection Profiles of the Italian Crust. Mem. Descrittive della Carta Geol. d’Italia, v. LXII.APAT, Dipartimento Difesa del Suolo, Roma, 194 pp.

Selvaggi G. and Amato A.; 1992: Subcrustal earthquakes in the Northern Apennines (Italy): evidence for a still activesubduction? Geophys. Res. Lett., 19, 2127-2130

Serri G., Innocenti F. and Manetti, P.; 1993: Geochemical and petrological evidence of the subduction of delaminatedAdriatic continental lithosphere in the genesis of the Neogene-Quaternary magmatism in central Italy.Tectonophysics, 223, 117-147.

Shapiro N.M., Ritzwoller M.H., Molnar P. and Levin V.; 2004: Thinning and flow of Tibetan crust constrained byseismic anisotropy. Science, 305, 233-236.

Sinigoi S., Quick J.E., Peressini G. and Mayer A.; 2003: An example of the Apulian lower crust: the Ivrea-Verbanozone. In: Transalp Conference Ext. Abstract, Mem. di Sci. Geol., Soc. Coop. Tipografica, Padova (Italy), 54, pp.101-104.

Tinivella U., Accaino F., Rossi G. and Nicolich R.; 2005: Petrophysical analysis of CROP-18 crustal seismic data. Boll.Soc. Geol. It., 3, 205-211.

Ursin B.; 1990: Offset-dependent geometrical spreading in a layered medium. Geophysics, 55, 492-496.

Velicogna I., Marson I. and Suhadolc P.; 1996: Morfologia della Moho da gravità, topografia ed isostasia. In:Bernabini M. (ed), Atti 14° Conv. GNGTS, Esagrafica, Roma, pp. 405-418.

Weiss T., Siegesmund S., Rabbel W., Bohlen T. and Pohl M.; 1999: Seismic velocities and anisotropy of the lowercontinental crust: A review. Pure Appl. Geophys., 156, 97-122.

Wilson C.K., Jone, C.H., Molnar P., Sheehan A.F. and Boyd O.S.; 2004: Distributed deformation in the lower crust andupper mantle beneath a continenatl strike-slip fault zone: Marlborough fault system, South Island, New Zealand.Geology, 32, 837-840.

Yilmaz O.; 2001: Seismic data analysis: processing, inversion and interpretation of seismic data. Series: Investigationin Geophysics, vol. 10. Society of Exploration Geophysicists-SEG, Tulsa (USA), 2077 pp.

Corresponding author: Flavio AccainoIstituto Nazionale di Oceanografia e di Geofisica Sperimentale - OGSDipartimento di Geofisica della Litosfera; Borgo Grotta Gigante 42/c, Sgonico (Trieste), ItalyPhone: +39 040 2140259; fax: +39 040 327521; e-mail: [email protected]

445

Highlighting the crustal structure of southern Tuscany by the … › bgta › pdf ›...

Documents

Transcript of Highlighting the crustal structure of southern Tuscany by the … › bgta › pdf ›...